Production ready LLM model compression/quantization toolkit with accelerated inference support for both cpu/gpu via HF, vLLM, and SGLang.

• 11/19/2024 1.2.2 HF GLM model support. ClearML logging integration. Use device-smi and replace gputil + psutil depends. Fixed model unit tests.
• 11/11/2024 🚀 1.2.1 Meta MobileLLM model support added. lm-eval[gptqmodel] integration merged upstream. Intel/IPEX cpu inference merged replacing QBits (deprecated). Auto-fix/patch ChatGLM-3/GLM-4 compat with latest transformers. New .load() and .save() api.
• 10/29/2024 🚀 1.1.0 IBM Granite model support. Full auto-buildless wheel install from pypi. Reduce max cpu memory usage by >20% during quantization. 100% CI model/feature coverage.
• 10/12/2024 ✨ 1.0.9 Move AutoRound to optional and fix pip install regression in v1.0.8.
• 10/11/2024 ✨ 1.0.8 Add wheel for python 3.12 and cuda 11.8.
• 10/08/2024 ✨ 1.0.7 Fixed marlin (faster) kernel was not auto-selected for some models.
• 09/26/2024 ✨ 1.0.6 Fixed quantized Llama 3.2 vision quantized loader.
• 09/26/2024 ✨ 1.0.5 Partial Llama 3.2 Vision model support (mllama): only text-layer quantization layers are supported for now.

GPTQModel started out as a major refractor (fork) of AutoGTQP but has now morphed into a full-stand-in replacement with cleaner api, up-to-date model support, faster inference, faster quantization, higher quality quants and a pledge that ModelCloud, together with the open-source ML community, will take every effort to bring the library up-to-date with latest advancements and model support.

Public tests/papers and ModelCloud's internal tests have shown that GPTQ is on-par and/or exceeds other 4bit quantization methods in terms of both quality recovery and production level inference speed in both token latency and rps. GPTQ has currently the optimal blend of quality and inference speed you would want to use in a real-world production system.

• 🚀 Extensive model support for: IBM Granite, Llama 3.2 Vision, MiniCPM3, GRIN-Moe, Phi 3.5, EXAONE 3.0, InternLM 2.5, Gemma 2, DeepSeek-V2, DeepSeek-V2-Lite, ChatGLM, MiniCPM, Phi-3, Qwen2MoE, DBRX (Converted).
• ✨ 100% CI coverage for all supported models including quality/ppl regression.
• 🚀 vLLM inference integration for quantized model where format = FORMAT.GPTQ
• 🚀 SGLang inference integration for quantized model where format = FORMAT.GPTQ
• 🚀 Intel/AutoRound QUANT_METHOD support added for a potentially higher quality quantization with lm_head module quantization support for even more vram reduction: format export to FORMAT.GPTQ for max inference compatibility.
• 🚀 Intel/IPEX support added for 4 bit quantization/inference on CPU.
• 🚀 BITBLAS format/inference support from Microsoft
• 🚀Sym=False Support. AutoGPTQ has unusable sym=false. (Re-quant required)
• 🚀lm_head module quant inference support for further VRAM reduction.
• 🚀 Faster quantization: More than 50% faster for TinyLlama + 4090 with batching and large calibration dataset.
• 🚀 Better quality quants as measured by PPL. (Test config: defaults + sym=True + FORMAT.GPTQ, TinyLlama)
• 🚀 Model weights sharding support
• 🚀 Security: hash check of model weights on load
• 🚀 Over 50% faster PPL calculations (OPT)
• 🚀 Over 40% faster packing stage in quantization (Llama 3.1 8B)

🤗 ModelCloud quantized ultra-high recovery vortex-series models on HF

GPTQModel is validated for Linux x86_64 with Nvidia GPUs. Windows WSL2 may work but un-tested.

# You can install optional modules like autoround, ipex, vllm, sglang, bitblas, and ipex.
# Example: pip install -v --no-build-isolation gptqmodel[vllm,sglang,bitblas,ipex,auto_round]
pip install -v gptqmodel --no-build-isolation
uv pip install -v gptqmodel --no-build-isolation

# clone repo
git clone https://github.com/ModelCloud/GPTQModel.git && cd GPTQModel

# pip: compile and install
# You can install optional modules like autoround, ipex, vllm, sglang, bitblas, and ipex.
# Example: pip install -v --no-build-isolation gptqmodel[vllm,sglang,bitblas,ipex,auto_round]
pip install -v . --no-build-isolation

Below is a basic sample using GPTQModel to quantize a llm model and perform post-quantization inference:

from datasets import load_dataset
from transformers import AutoTokenizer
from gptqmodel import GPTQModel, QuantizeConfig

model_id = "meta-llama/Llama-3.2-1B-Instruct"
quant_path = "Llama-3.2-1B-Instruct-gptqmodel-4bit"

tokenizer = AutoTokenizer.from_pretrained(model_id)

calibration_dataset = [
  tokenizer(example["text"])
  for example in load_dataset(
    "allenai/c4",
    data_files="en/c4-train.00001-of-01024.json.gz",
    split="train"
  ).select(range(1024))
]

quant_config = QuantizeConfig(bits=4, group_size=128)

model = GPTQModel.load(model_id, quant_config)

model.quantize(calibration_dataset)

model.save(quant_path)

model = GPTQModel.load(quant_path)

result = model.generate(
  **tokenizer(
      "Uncovering deep insights begins with", return_tensors="pt"
  ).to(model.device)
)[0]

For more advanced features of model quantization, please reference to this script

Read the gptqmodel/models/llama.py code which explains in detail via comments how the model support is defined. Use it as guide to PR for to new models. Most models follow the same pattern.

GPTQModel inference is integrated into lm-evaluation-hardness and we highly recommend avoid using PPL and use lm-eval to validate post-quantization model quality.

# currently gptqmodel is merged into lm-eval main but not yet released on pypi
pip install lm-eval[gptqmodel]

• GPU: Marlin, Exllama v2, Triton kernels in that order for maximum inference performance. Optional Microsoft/BITBLAS kernel can be toggled.
• CPU: Intel/IPEX kernel

@misc{gptqmodel,
    author = {ModelCloud.ai},
    title = {GPTQModel},
    year = {2024},
    publisher = {GitHub},
    journal = {GitHub repository},
    howpublished = {\url{https://github.com/modelcloud/gptqmodel}},
}

@article{frantar-gptq,
  title={{GPTQ}: Accurate Post-training Compression for Generative Pretrained Transformers}, 
  author={Elias Frantar and Saleh Ashkboos and Torsten Hoefler and Dan Alistarh},
  year={2022},
  journal={arXiv preprint arXiv:2210.17323}
}

@article{frantar2024marlin,
  title={MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models},
  author={Frantar, Elias and Castro, Roberto L and Chen, Jiale and Hoefler, Torsten and Alistarh, Dan},
  journal={arXiv preprint arXiv:2408.11743},
  year={2024}
}